Caching and Reproducibility: Making Data Science Experiments Faster and FAIRer.

Small to medium-scale data science experiments often rely on research software developed by individual scientists or small teams. Often there is no time to make the research software fast, reusable, and open access. The consequence is twofold.